Screening for mouse genes lost in mammals with long lifespans
Background: Gerontogenes include those that modulate life expectancy in variousspecies and may be the actual longevity genes. We believe that a long (relative tobody weight) lifespan in individual rodent and primate species can be due, amongother things, to the loss of particular genes that are present in short-lived species ofthe same orders. These genes can also explain the widely different rates of agingamong diverse species as well as why similarly sized rodents or primates sometimeshave anomalous life expectancies (e.g., naked mole-rats and humans). Here, weconsider the gene loss in the context of the prediction of Williams’ theory thatconcerns the reallocation of physiological resources of an organism between activereproduction (r-strategy) and self-maintenance (K-strategy). We have identified suchlost genes using an original computer-aided approach; the software considers theloss of a gene as disruptions in gene orthology, local gene synteny or both.Results: A method and software identifying the genes that are absent from apredefined set of species but present in another predefined set of species aresuggested. Examples of such pairs of sets include long-lived vs short-lived,homeothermic vs poikilothermic, amniotic vs anamniotic, aquatic vs terrestrial, andneotenic vs nonneotenic species, among others. Species are included in one of twosets according to the property of interest, such as longevity or homeothermy. Theprogram is universal towards these pairs, i.e., towards the underlying property,although the sets should include species with quality genome assemblies. Here, theproposed method was applied to study the longevity of Euarchontoglires species. Itlargely predicted genes that are highly expressed in the testis, epididymis, uterus,mammary glands, and the vomeronasal and other reproduction-related organs. Thisagrees with Williams’ theory that hypothesizes a species transition from r-strategy toK-strategy. For instance, the method predicts the mouse gene Smpd5, which has anexpression level 20 times greater in the testis than in organs unrelated toreproduction as experimentally demonstrated elsewhere. At the same time, itsparalog Smpd3 is not predicted by the program and is widely expressed in manyorgans not specifically related to reproduction.Conclusions: The method and program, which were applied here to screen forgene losses that can accompany increased lifespan, were also applied to studyreduced regenerative capacity and development of the telencephalon, neoteny, etc.Some of these results have been carefully tested experimentally. Therefore, weassume that the method is widely applicable.